Re: limitation for macro expansion

[Yasushi SHOJI <yashi@atmark-techno.com>]

Hi Nicolas,

On Mon, Mar 6, 2017 at 5:22 PM, Nicolas Goaziou <mail@nicolasgoaziou.fr> wrote:
> Yasushi SHOJI <yashi@atmark-techno.com> writes:
> there are two ways to interpret it: the one you expect and
>
>   a_{CONTENTS} where CONTENTS is {{something}}.
>
> Since subscript syntax started first, it has precedence over the macro.

Yes.  That's what I understand from your response.  Thanks!

>> Doesn't seems to work.  When I export 'a_{{{{author}}}}', it
>> generates:
>>
>>   a_{Yasushi SHOJI}
>>
>> I was expecting to see:
>>
>>   a_Yasushi SHOJI
>
> This is correct if you use `ascii' back-end.

Sorry for the confusion.
I just thought that you gave me a way to use macro with underscore.

>> Would it be OK to say that it's a design decision to ignore the macro
>> expansion in the link field at export time?
>
> It is. Actually, macros are not allowed in many places, as a design
> decision. See
>
>   (info "(org) Macro replacement")
>
> for more details.

I assume that the key phrase is "anywhere Org markup is recognized".
Link format doesn't allow Org markup, right?

>> ie)
>>
>>   #+MACRO: timestamp {{{date(%Y%m%d)}}}
>>
>>   Please open log_{{{timestamp}}}.txt
>>
>> Would it be possible for org to do this?
>
> You could try
>
>   #+MACRO: timestamp {{{time(%Y%m%d)}}}
>
>   Please open log\under{}{{{timestamp()}}}.txt

Ah!  Great idea.  That's what I'm looking for. Thanks.

# I use file_name_with_underscore.txt more than subscripts
# I'd be nice, at least for me, to have '\sub' and '\super' special keywords
# but leave the underscores alone.

>> If not, would it be possible for me to modify the code to achieve this?
>> My stupid idea is to:
>>
>>   - disable sub / superscript parser when org-use-sub-superscripts is nil
>
> See `org-export-with-sub-superscripts'.

hmm.  just checked the source.  org-use-sub-superscripts is only for display.
org-export-with-sub-superscripts is just for exports.

I was gonna just by-pass or disable subscript parser all together when
org-use-sub-superscripts is nil but it doesn't seems to be a good idea, does it?

>>   - reverse the precedence order of subscript and macro
>
> This would only move the problem elsewhere.

True.
-- 
             yashi